Advanced computational tools for PEM fuel cell design Part 1. Development and base case simulations

This paper reports on the development and numerical implementation of a comprehensive 3D computational fuel cell dynamics code for PEMFCs. The code solves a set of coupled non-linear conservation equations (mass, momentum, species, energy, electrical potential and liquid water saturation) for an ent...

Full description

Saved in:
Bibliographic Details
Published inJournal of power sources Vol. 180; no. 1; pp. 410 - 422
Main Authors SUI, P. C, KUMAR, S, DJILALI, N
Format Journal Article
LanguageEnglish
Published Lausanne Elsevier Sequoia 15.05.2008
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:This paper reports on the development and numerical implementation of a comprehensive 3D computational fuel cell dynamics code for PEMFCs. The code solves a set of coupled non-linear conservation equations (mass, momentum, species, energy, electrical potential and liquid water saturation) for an entire unit cell. A phenomenological model for water transport in the membrane is solved separately for the membrane domain, in conjunction with calculation of the water content on the boundary such that that water balance is satisfied on both sides of the membrane interface, and the numerical implementation of the model is validated against an analytical solution. The global polarization curve predicted with the CFD code is found to compare favorably with reported data. A detailed validation of the CFD code against spatially resolved experimental data is presented in a companion Part 2 paper, and in this paper base case simulations for a unit cell with straight channels are presented to illustrate and analyze basic physical features, transport of species along the channel and coupling between heat and mass transfer processes. Analysis of the results shows that many of the variables of interest, including mass fractions and current densities, exhibit similar profiles along the channel, which suggests that reduced dimensional model based on appropriate similarity variables might be suitable for rapid calculations.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0378-7753
1873-2755
DOI:10.1016/j.jpowsour.2008.02.007